© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 62544
    Prev
    Next
    INDEX
    Explanations

    2

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    éĻĤ
    -0.28
    ExceptionHandler
    -0.26
    æ¸Ĭ
    -0.25
     yapıl
    -0.25
     BASIS
    -0.25
    wert
    -0.25
     orientations
    -0.24
    ordes
    -0.24
    Trash
    -0.24
     scopes
    -0.24
    POSITIVE LOGITS
    èµĽåIJİ
    0.29
    udget
    0.26
    ument
    0.26
    åĽĽæĺ¯
    0.26
    æĸ¯é¡¿
    0.24
    èĪ·
    0.24
    anta
    0.24
    rus
    0.24
    æ¦Ĩ
    0.23
    Ĥ¬
    0.23
    Activations Density 88.162%

    No Known Activations